The AM Forum

any ideas on tetrode screen regulation

I've never heard of a tetrode that needs 2kv on the screen grid...What the heck is the plate voltage?  ??? What kind of tube is this?

Also what class of operation, and mode are you using it for? Does the existing screen supply work, but performance suffers in some way wanting you to regulate the screen supply?

Is this rig in 27.185 Mhz pulse mode? I always wanted to do that.... ;D

Jim
WD5JKO

Go to Richard Measure's website. He has an adjustable 2KV MOSFET regulated supply. Pretty simple.  You don't want the screen to sag at all, thus get rid of that unregulated supply situation.  Any unreg supply going from ~50 to 350 ma is gonna sag and cause distortion, splatter, etc to some degree.

A stiff screen voltage is mandatory for clean tetrode linear amplifier operation.

T

Last summer I read an article that someone is making a 4 KV IGBT. This would make a nice series pass device. Just need to bias the gate with a zener stack.
Price was around $60 at Digi Key.

sorry meant 500 volts on the screen

class ab1

As Steve McQueen said (breathlessly) in the Great Escape,   WoW!  I am sure it will provide plenty of output for AM at our limit, say forever.

?????.

Sell it to a CBer and buy what you want.  I am sure there are some that will pay big bucks to use it mobile in their vehicle.

I like home brew

T

Henry made great amplifiers

One thing commercial installations do is maintence and that is the thing amateurs don't do on a schedule.  If someone would clean the inductors and contact assembly, they might not have that problem.  Have seen many Henry amps with rollers that were properly maintained and they did not do that. 

One time per year, on a regular schedule, pull the equipment apart and inspect it, clean it and make any needed adjustments.

I agree with you on that

Dont get me wrong i think henry made great amps .

Henry, like Collins and all other manufacturers, built their equipment to match a certain impedance.  If it was tuned out of that range or near its limits anything can happen.  Abuse by a former owner is no reason to criticize the construction, especially if it was operated outside of it design specifications, which you seem to indicate.

Now if you want an amp that will load into the bedsprings at over 2 KW you seem to have one.

your right

Why not machine your own rollers out of silver.

would still have to change the shafts they ride on.
Plus the arc problem would still be there.

If i had one i would make it work but wouldnt pay 10-12k for a unit that needs to be repaired or even a working one.

sounds kind of like a 208U-10.

I have a Henry 8K on the bench right now for repair, what a Rube Goldberg POS!

One owner since 94, runs it legal and yet its fried 1 tube, a 24V PS, a bunch of stuff in the tank switching and he doesnt want to send it back on a 6000 mile RT again.

This time the input VSWR went high yet everything tests fine. The 3CX3000 idles normally, just wont take drive.

Carl
KM1H

As much as I like Rich Measures work I have to say I don't like that screen regulator much, for LOTS of reason I won't get into. But mainly it won't deal with negative screen current as already pointed out.  If you already have pretty stiff regulation now I'd leave well enough alone. If you insist on going to a regulated supply then a shunt regulator would be my choice. 

Microwave oven capacitors make good resonating capacitors...... Most of the values seem just about right for resonating chokes at 100hz, 120 hz would be simply a matter of extra capacitance. I use a scope to tune the choke, just make sure your probe has at least a 400% safety margin, the voltages can soar to a very high value!

Whats wrong with the Measures screen regulator, I am curious? I thought that it could handle negative screen current? I built one for a 4-1000 and never had any problems, although I cant recall measuring negative screen current. It has very high value of bleeder current.

I cant  see how you going to build a  shunt regulator with such high screen dissipation,  high wattage zeners are hard to find and very expensive..  I am willing to learn on this point.

I certainly would not want to run any AB1 tetrode without a regulator screen supply!!

Heres a very interesting screen regulator. I am sure you could  size it up.

http://www.nd2x.net/JH0WJF-2.html

Craig

"Heres a very interesting screen regulator. I am sure you could  size it up."
http://www.nd2x.net/JH0WJF-2.html


Craig,

   The original need here is a screen regulator to provide 1500-200v at up to 350 ma. Doing this with a solid state regulator would certainly be challenging. The circuit you link to would be difficult to scale up. I also see that the loop gain is low, and therefore the no load to 10K loaded output voltage has significant sag, maybe more than a good choke input filter with a good bleeder load..

   If I were to regulate this screen supply at 1500v-2000v @ 0-350ma, I might use a 304TL tube in a shunt regulator configuration. I might actually need two of them if the 300w dissipation is challenged.  ???

Jim
WD5JKO








[/quote]

Craig,
   Are you saying that you have a bleeder to ground from the screen? That’s not shown or mentioned in the design of this screen regulator. But if that’s the case it’s certainly a big improvement, even though screen voltage will still raise some.
   Since you ask… and in lieu of an appropriate screen bleed resistor…
   The high voltage regulator that Rich Measures posted was developed from the manufacturer’s app. notes for the LM723 as a generic high voltage supply. Unless I missed something it appears it was not specifically re-designed to regulate a load which could also become a current source. Hence if the screen were to go positive there is nowhere for that current to sink to hold down the voltage. 

   Next, the “crowbar” as it’s called seems to protect the pass transistor from excess current (over-voltage actually) but does nothing to protect the load should it short.  If the screen were to short the regulator is incapable of shutting down or current-limiting the screen current.  R1 limits current to protect the crowbar? A kilovolt regulator limited by 1 ohm?  Seems we have the protected vs. the protector circuits confused here. Screw protecting the regulator, what about my expensive tube!?
   The regulator requires a separate floating LV supply capable of withstanding the screen voltage. Seems like a lot of unnecessary parts to me. 
      I’m also a little skeptical of the ground reference being stable under varying loads. R2 is a pretty high value to 0 ground.  It has screen voltage + reference voltage across it. The reference uses a voltage source local to the regulator but employs ground in its return path to establish the reference voltage across R2, any fluctuation across R2 would not be compensated for by the regulator.  No other path for the reference current seems apparent except through the screen itself.  I’d rather not employ my load current as part of any reference if avoidable.  This is a significant problem in the event the screen goes positive.  .
   R5/R6 establish the regulator’s gain and sensitivity As a 2:1 divider the sensitivity and stability are very good but relies heavily on Q2 for the voltage gain required in this application.  Not a bad thing, merely pointing it out.  It seems the 723’s features are not being used to their full potential.
   The link you provided is certainly a preferable design, IMHO, being a shunt regulator. I do not like the resistance in series with the regulating element since that greatly decreases its ability to stabilize the screen voltage.    I agree with Jim, W5JKO,  that a more robust shunt device would be preferable. Positive screen current is then limited by supply resistance and negative current has a low impedance path to ground. 
Regards,
Mark
   

Hi Jim

The 1968 handbook has  a screen regulator design for 1500 volts. It uses a 807.

Before I used  the Measures screen regulator I used a resonant choke supply for the 4-1000. I could not Measure any IMD improvement when I switched over to the regulated Measures screen regulator


Mark  KA2QFX

Thanks for your appraisal of the design.  I have had no problems whatsover with his screen regulator design.  If you look at the grid circuit  that Rich has on his web page you can see the resistor and protection diodes. 

http://www.somis.org/D_a_05.GIF

On my 4-1000 I am using a 40 k  resistor directly across the output  of the screen supply. I am also using a grounded screen.
I have  had many screen glitches and even a dead  short, the voltage has always remained regulated. What this circuit is lacking, is screen over current protection. I just used a simple over current relay trip.   



73
Craig

Seeing the complete application of that screen regulator makes a significant difference in my assessment. But series regulation still wouldn't be my preference.

Nice tip on the microwave oven caps. I have a few of those. I'll have to check out their value versus some of the chokes I have.

Mark

After some thought, a 1500-2000v @ 0-350ma screen supply, will be a challenge to do with regulation.

In my opinion, doing this solid state is out, unless someone really wanted to do that...If I did it, I'd use a triac on the transformer primary to vary the voltage in a near lossless manner.

For a shunt regulator approach, we need to sink 350ma when the screen current is 0ma, so at 2KV out, 350ma sinked, the shunt regulator dissipates 700 watts! I mentioned a 304TL, but now I realize it world take three of them! Or how about a 4-1000, that would do?

The series resistor between the supply and the shunt regulator would also have to dissipate a bunch of watts also.

So what's wrong with a series regulator instead? Well the negative screen current that might occur with a tetrode is the concern. So how about a series regulator followed by a bleeder resistor? Make the bleeder pass 50ma at 2KV. This would mean that the series regulator stays in regulation until the negative screen current rises to 50ma. Above that, we go out of regulation. So how much negative screen current occurs on this big tetrode?

Early on in this thread, it was said the AMP was for AM linear operation. If so, then the screen current will be constant with/without modulation. Therefore regulation is not a big issue, or is it?

Jim
WD5JKO

Jim (et al)

Agreed, 0-350mA at 1.5~2KV is quite a lot to handle.  Under the circumstances you point out, given that negative screen current would not likely be much, series regulation with a sizable bleeder would seem the lesser of all evils.

But I can't help but think that if really stiff regulation is a priority perhaps a push-pull regulation scheme could be employed where neither supply nor shunt element need pass excess current unless required. As long as the unregulated supply voltage isn't too high dissipation could be held to a minimum.   

As to the need of a REALLY stiff screen supply...
I've always considered the screen voltage primarily as a method of establishing the Mu of the tube, and accordingly, its saturation level, plate to grid leakage, etc.  If allowed to fluctuate sufficiently this could be seen as a considerable source of non-linearity. The G2DAF amplifier design deliberately took advantage of just this trait.  :P  However, I do not consider a less than stiff screen supply as a very significant cause of non-linear distortion.
Rather, I see improper bias a much more significant contributor to distortion products. I have always found that a higher than desirable bias usually yielded the most favorable outcomes for cleanliness, with or without really stiff screen voltages. Of such factors are compromises determined between efficiency and cleanliness. 

I know of a few grid driven tetrodes of this power class...  Most users run them strapped as a triode, but I prefer the ease of drive of the tetrode configuration.

All that I know of just use the bleeder R as the "regulator".  This is in AM service, usually.

My big amp, a pair of 4X15s, is run as a pair of triodes now.  The IPA is a 4CX5000 grid driven.  I use an R across the screen supply, works FB thus far (going on a decade).

You're right about the filament inrush being the main reason they recommend power cycling once / day.  I variac my filament xformer, as well as float it above chassis ground.  Pretty hard to built a 350A filament choke, much easier to do it on the primary side.  Using a variac on the input to the filament choke only means having to remember to drop it to 0 every time you shut the amp off... Eventually it becomes second nature.

What about using a zener as a clamp for -g2?

--Shane